Forecasting and policy analysis system
The Macroeconomics and Statistics department draws the forecast along with the members of the Monetary Policy Committee. The forecast is based on macroeconomic modeling, which is supplemented with the professional views of NBG’s experts.
Implemented by the NBG team of economists, the forecasting process lasts about a month and involves meetings at the highest level of the management. The starting conditions of the forecast are discussed at the initial issue meetings. In particular, domestic and external economic environments are analyzed and the assumptions about exogenous factors and alternative scenarios regarding the expected development of the external sector are discussed. In this process, the NBG uses the forecasts of international institutions such as the International Monetary Fund (IMF), the World Bank, Bloomberg, etc. The initial stage of the forecasting process is essential for the final results.
In addition to assessing the initial conditions, the issue meetings also provide an analysis of the short-term forecast, which focuses on economic developments in the upcoming 1-2 quarters. The short-term forecast is based on statistical approaches to analyze empirical data and the expert judgment of the NBG’s professionals. At the next stage, the short-term forecast is incorporated into the macroeconomic model, which depicts the medium-term dynamics of the economy.
The NBG’s forecasting and monetary policy analysis system incorporates several analytical instruments, which could be conventionally classified into three categories: short-term forecasting tools, basic macroeconomic model (the core of the system), and additional satellite models.
Short-term forecasting instruments - In the short run, economic variables are characterized by inertia. Consequently, the application of econometric models for short-term forecasting is expedient. The National Bank of Georgia forecasts one or two quarters of inflation and real GDP growth by using the Error Correction Model (ECM), ARIMA and the Factor Model. At the same time, the official data on real GDP is published with a time lag. Thus, GDP growth of previous quarter is nowcasted with this tool. The short-term forecasts, in turn, are provided to the main macroeconomic model and are applied in the medium-term forecasting as the starting condition for a longer-term forecast.
The macroeconomic model, which is a main part of FPAS, is a semi-structural model. On the one hand, it is balanced by desired empirical properties and, on the other hand, by the Dynamic Stochastic General Equilibrium (DSGE) approach. It includes the microeconomic foundation and market expectations, making the model future-oriented. The model’s equations are structural and each of them is meaningful in terms of economic theory. The model was designed to meet modern policy requirements and to incorporate the special characteristics that fit the Georgian economy. In addition to the forecasting, using this model allows to discuss and analyze baseline and alternative scenarios, study the impact of different exogenous shocks on the model’s outcome, analyze different risks, perform counterfactual analysis, etc. Noteworthy, the structure of the model facilitates its constant development and improvement.
Satellite model - This model predicts the contributions of expenditure components to real GDP growth.
The macroeconomic model’s forecast endogenously takes into consideration the NBG’s monetary policy as one of the factors of the macroeconomic environment. Accordingly, the forecast of the model reflects the current and expected stance of the NBG. The monetary policy reaction function takes the form of the simple Taylor-type rule and is in line with the objectives defined in the Organic Law of the National Bank.
Detailed description of Forecasting and Policy Analysis System - see link